화학공학소재연구정보센터
Macromolecules, Vol.29, No.5, 1399-1406, 1996
Polymerization of Epsilon-Caprolactone Initiated by Aluminum Isopropoxide Carried Out in the Presence of Alcohols and Diols - Kinetics and Mechanism
The mechanism of polymerization of epsilon-caprolactone (epsilon CL) initiated with aluminum isopropoxide (Al((OPr)-Pr-i)3) trimer (A(3)) and/or tetramer (A(4)) carried out in the presence of ethanol, 2-propanol, 1,2-ethanediol, 1,5-pentanediol (PD), and poly(ethylene glycol) as chain-transfer agents was studied. It has been established, by the kinetic studies, that the low-molecular-weight alcohols and diols, introduced purposely into the polymerization mixture, operate not only as the chain-transfer agents but also either as inhibitors, in polymerization initiated with A(3), or as accelerators, in polymerizations initiated with A(4). Thus, the kinetic curves are practically identical for different compositions of the A3/A(4) mixture in the feed, under the otherwise comparable conditions. At the early stages of polymerization the relative polymerization rate (r(p) = -d ln [epsilon CL]/dt) increases, and at higher conversions, after complete consumption of the starting alcohol or diol, r(p) becomes constant; i.e. polymerization becomes first-order (internally) in monomer. These kinetic effects were explained by assuming the complete disruption of any aggregate present, followed by a coordination of the resulting nonaggregated (monomeric) Al-tris(macroalkoxide) active species with the low-molecular-weight alcohol or diol. Analysis of the H-1 and Al-27 NMR spectra of the polymerizing mixtures confirmed that both A(3) and A(4) are consumed quantitatively. This is in contrast to the polymerization conducted without alcohols, in which only A(3) is reacted fast, whereas A(4) mostly remains unreacted when the polymerization is over. Moreover, the H-1 NMR spectra show that the equilibrium > Al-(OPr)-Pr-i + HO(CH2)(5)OH reversible arrow > Al-O(CH2)(5)-OH + (PrOH)-Pr-i is shifted to the right-hand side. The kinetic scheme taking into account both propagation, P-n* + epsilon CL --> P-n+1*, and the temporary deactivation of the growing species, P-n* + PD reversible arrow P-n*. PD (where the complex P-n*. PD does not propagate), was proposed. A set of the resulting differential equations was solved numerically. The computed kinetic plots, ln([epsilon CL]/[epsilon CL]) and [PD] vs time, reproduce well the respective plots obtained experimentally.